Weird statistical reporting from Rasmussen Reports

For reasons unfathomable to me, the US public opinion poller Rasmussen Reports has decided to measure President Obama's popularity by a Presidential Approval Index rating arrived at by strange means. They get it by subtracting the percentage of those who strongly disapprove of his Presidency from those who strongly approve of it.

Why exclude moderate approval and disapproval? The current figure of -8% arrived at by Rasmussen Reports through their methodology seems to the casual viewer to suggest a negative overall view of President Obama. Yet the total numbers provided by Rasmussen in the article (50% overall approval and 49% overall disapproval) suggest a much more even split. I don't see much room here for interpreting this as Obama having negative overall popularity the way this Index misleadingly suggests. I think that there is room for interpreting this as Obama being more unpopular among rabid partisans than he is among the general population, though.

Curious also is the decision by Rasmussen to only survey likely voters to determine approval rating. As the survey summary notes, "President Obama's numbers are always several points higher in a poll of adults rather than likely voters". I can understand why this might be relevant in a poll of how people are likely to actually vote, but in general overall approval? Does being unlikely to vote automatically make your opinion completely worthless when it comes to politics? I can see how some people might argue that it should, but personally I think the principle of democratic accountability extends well beyond the single moment of an election. If people, even people unlikely to vote, have an opinion about government, then that opinion should be known and taken into account. It's a shame that Rasmussen doesn't do that.

It does seem awfully convenient that both these statistical oddities have the effect of making President Obama's approval rating seem much lower than the results reported in other surveys with more meritorious methodologies. If I was a conspiracy-oriented person I might start wondering if this was deliberate. Good thing I'm not a conspiracy-oriented person, then.

21 Reasons Why Gender Matters: Paul Cameron references

I go back to university soon, so I won't have much time to write here. I didn't get through nearly as many of the references of the local anti-gay pamphlet "21 Reasons Why Gender Matters" as I would have wanted over the holidays, but realistically I suppose it's too tough a job for one person. There are still a few things I have found, though. The most significant would probably be the use of not one, but two, direct references from Paul Cameron, an "expert on homosexuality" so discredited that there are even anti-gay groups who consider his work too shoddy and unreliable to use (and that's saying something).

The first is using his recent distortions about the average lifespan of gay and lesbian partnered people in Europe:

Scandinavian research has shown that married homosexuals’ and
lesbians’ life spans are 24 years shorter than heterosexual couples.
In Denmark over the 12 years after 1990, the average age of death
of hetero men was 74, whereas the 561 partnered homosexual men
who died in the same period did so at an average age of 51. Married
women died at an average age of 78, whereas the nine lesbian women
who died, did so at an average age of 56. In Norway the figures were
similar – married heterosexual men died at an average age of 77, the
31 homosexuals at 52; heterosexual women died at 81, while the 6
lesbians who died, did so at mean 56.151

With footnote 151 being none other than

151 Paul Cameron,. “Federal Distortion of The Homosexual Footprint.”:

which has already been debunked in detail by Box Turtle Bulletin.

The second is the usual conflation of homosexuality and child molestation

A recent review of the child molestation literature as it appears in
medical and psychological journals concluded that between 25 and 40
per cent of all recorded child molestation was homosexual.169

With footnote 161 being

169 Reported in Paul Cameron, Exposing the AIDS Scandal. Lafayette, Louisiana: Huntington House Publishers, 1988, p39.

Child molestation is child molestation of course, and it's only anti-gay activists who falsely treat molesters of boys and girls as subsets of the larger heterosexual and homosexual population. But I am somewhat amused by the description by the Gender Matters pamphlet of a Cameronite writing from 1988 as a "recent review of the child molestation literature". Getting desperate for relevance, these anti-gay types are.

21 Reasons Why Gender Matters: Footnote 66 is another mistake

The pamphlet "21 reasons why gender matters" can be found online here.

Footnote 66 is the alleged source of these statistics on the length of male same-sex relationships:

a study of the Melbourne homosexual community showed that 40 per cent of men had changed partners in the past 6 months; 9.8 per cent had been in a relationship for only six months to a year; 18.8 per cent for 1-2 years; 15.3 per cent had lasted for 3-5 years; and only 15.7 per cent were in a relationship of more than five years – meaning 84 per cent had broken down after five years.

They supposedly come from the "Melbourne Gay Community Periodic Survey, 2000" by Clive Aspin et al, published by the National Centre in HIV Social Research.

The survey results are available online. It is simply impossible for the statistics attributed to it to have come from there. The Periodic Survey only measures whether the regular relationships of those surveyed have lasted for "less than 1 year" or for "at least one year" (see Table 9 on page 12 of the Periodic Survey). Even the most statistically illiterate person around would have trouble getting the Fatherhood Foundation's alleged statistics from the data actually in the Periodic Survey: in 2000, 31.8% of those in a regular relationship had been in it for less than a year, while 68.1% had been in it for a year or more.

The website of the Australian Christianist group the Saltshakers includes a statistics page on relationships which quotes the same numbers but gives a different source, although it's easy to get confused (which the Fatherhood Foundation apparently did. Again). Their source for the numbers is "Men and Sexual Health", by the National Centre in HIV Social Research, 1997. There doesn't appear to be any kind of study with this name. Are they referring to the longitudinal cohort study called "Sydney Men and Sexual Health" (SMASH)? It's hard to say, as the main report that I can find on that study is a book called "Methods and sample in a study of homosexually active men in Sydney, Australia" that was, er, published in 1995. The book is accessible to me, once university libraries end their holiday closing period next week.

So the Fatherhood Foundation has, for the third time that I've now found, supplied a reference in their pamphlet "21 Reasons Why Gender Matters" that is verifiably wrong. I hope people keep that in mind when anti-gay activists rabbit on about how "well-referenced" this little smear pamphlet supposedly is. And the only alternative source given for the quote - from another anti-gay organisation - does not appear to be correct either. Maybe I'll be lucky enough to find the true source next week, but we'll see.

By the by, the reason that anti-gay groups continue to get away with telling these lies is because nobody effectively challenges them. That's one of the reasons I continue to hunt these misquotes down despite the obstacles and frustrations that their inaccurate referencing throw up. I hope it proves useful to somebody someday.

Statistical skullduggery from the Fatherhood Foundation: "proving" gay relationships shorter

On page 10 of the anti-gay pamphlet "21 reasons why gender matters" there is a piece of statistical deception that is very popular among anti-gay circles. It is one that needs no examination of references to spot.

The text is as follows:

Heterosexual married couples have a far lower rate of relationship breakdown than homosexual couples. As an Australian Government report stated, “According to a 1995 study, ten per cent of marriages failed within six years, 20 per cent within ten years, 30 per cent by twenty years, and 40 per cent by thirty years.”65 In comparison, a study of the Melbourne homosexual community showed that 40 per cent of men had changed partners in the past 6 months; 9.8 per cent had been in a relationship for only six months to a year; 18.8 per cent for 1-2 years; 15.3 per cent had lasted for 3-5 years; and only 15.7 per cent were in a relationship of more than five years – meaning 84 per cent had broken down after five years.

The lie is glaringly obvious to anyone within even a slight understanding of statistics: the Fatherhood Foundation is comparing every single homosexual relationship to only those heterosexual relationships that are called "marriage". They've deliberately and dishonestly weighted the heterosexual side of the comparison by excluding all unmarried opposite-sex couples, and then used that false data to try and paint homosexual couples as inherently inferior across the board. Unmarried opposite-sex couples, and not married opposite-sex couples, would be the real equivalent of unmarried same-sex male couples such as those allegedly studied in Melbourne. But making a comparison that's actually valid wouldn't give the Fatherhood Foundation the opportunity to smear the gay community with their misinformation now, would it?

21 Reasons why Gender Matters: shonky referencing and the mental health of gay people

I found a second serious referencing error in the anti-gay pamphlet 21 Reasons why Gender Matters. Like the earlier one, it falsely claims that a study includes text which it doesn't actually include. Unlike the earlier one, I can actually see how they got it wrong.

On page 14 of the pamphlet, the following text appears:

One study revealed that “the lifetime prevalence for two or more psychiatric disorders for men who engaged in homosexual behaviors was 37.85 per cent versus 14.4 per cent for men who did not engage in homosexual behaviors. For women engaging in homosexual behaviours, the rate for two or more psychiatric disorders was 39.5 per cent versus 21.3 per cent for women not engaging in homosexual behaviours. Society’s oppression of homosexual people is a hypothesis unlikely to find support in this study, concluded the Netherlands [sic], which is perhaps one of the most homosexual-affirming and tolerant countries in the world.”110

Footnote 110 refers to a study that is very popular among anti-gay activists trying to "prove" that homosexual behaviour itself directly causes the person engaging in it to become mentally disturbed: "Same-sex Behavior and Psychiatric Disorders", by TGM Sandfort et al, published in volume 58(1) of the Archives of General Psychiatry in 2001.

I have now read the complete study, and the text quoted above as appearing in the study itself does not appear anywhere within in. Given the actual sentiments expressed by the authors in the study, particularly their fairly clear statement in the "Comment" section that "because of the study's cross-sectional design, it is not possible to adequately address the question of the causes of the observed differences" in mental health, it is highly misleading to claim that they made any statement of fact as clear-cut as the one that the pamphlet falsely attributed to them.

The actual source of the quoted text is this article from the so-called "research and therapy" group NARTH, a little way in to "Section D:Mental Health, Physical Health, Stability of Homosexual Men and Women and Longevity of Homosexual Relationships". The shoddy use of HTML, in which a separate font is used when quoting a study, but that font accidentally spills out of the closing blockquote, could, if you're not reading carefully, give the misleading impression that the paragraph after the quote from the study is also a quote from the study. Apparently the Fatherhood Foundation didn't notice the problem.

The study itself that they're misquoting is interesting, and probably deserves a more thorugh consideration given the multiple lies that a great deal of anti-gay organisations tell about it, not just NARTH and the Fatherhood Foundation. For now I'll just point to a news article about a much more recent study about the issue of homosexuality and mental health, one which should give pause to any anti-gay activist trying to encourage negative attitudes to homosexuality: Parents' response key to health of gay youth:

Kids with parents who reacted negatively 8 times more likely to try suicide
by Lisa Leff
Associated Press
SAN FRANCISCO - Young gay people whose parents or guardians responded negatively when they revealed their sexual orientation were more likely to attempt suicide, experience severe depression and use drugs than those whose families accepted the news, according to a new study.

Follow the link for the full article.

21 Reasons Why Gender Matters: Footnote 2

Footnote 2: inaccurately listed as a second "footnote 1" in the main text. It's purportedly a source for the claim that "some people" refer to a condition called "gender disorientation pathology" (by which they mean "homosexuality"). It isn't a real academic footnote, saying only "refer to Section 15" of the exact same pamphlet.

Section 15 of the pamphlet concerns homosexuality and is predominantly, but not completely, unfootnoted pseudoscientific gobbledygook like "Many homosexuals report that as children, they had a dysfunctional relationship with their same-sex parent - such relationships being their primary means of gender identification and affirmation". Those limited areas which were footnoted in that section will, hopefully, be dealt with in time.

Verdict on Footnote 2: DISQUALIFIED DUE TO BEING SELF-REFERENTIAL

PS Failing to footnote correctly by repeating "footnote 1" twice in the text? Sloppy.

21 Reasons Why Gender Matters: Footnote 1

Well, this is going to get old really quickly, but I said I'd do it. I'm going to attempt to go through each footnote of the pamphlet 21 reasons why gender matters and check their validity. Having already seen 1 outright lie buried in footnote 82, I'd like to go through as many as I can for as long as I can. Can I do all 178? Probably not, but I'll attempt it.

To start with, Footnote 1: used to support the claim that "The great majority of single-parent families are fatherless". The footnote reads "In 2006, 87% of one-parent families with children under 15 years were headed by mothers. “Australian Social Trends, 2007: One Parent Families.” Canberra: Australian Bureau of Statistics, 2007."

Original source is here. The footnote's characterisation of the data is accurate.

However, the main text of the pamphlet's characterisation of this data is not. Note that the claim is that children in these families "grow up fatherless". However, according the ABS study quoted, the majority of single-parent families are created from divorce. A further substantial proportion are created from the break-up of de facto couples. Therefore the children from these families have had fathers in the lives, and in some cases may still do so, albeit only in the form of the father's visitation rights. They are not "fatherless" in the alarmist sense that the pamphlet claims.

This is not the evidence of masculinity in crisis that the pamphlet authors claim it is. It is evidence of the unfortunate prevalence of divorce, and possibly an argument that courts in custody battles side with the mother of a child much more than the father (a VERY common complaint of fathers in custody battles, by the way). But the conclusions drawn by this pamphlet here are misleading.

Verdict on Footnote 1: HALF-TRUTH

"21 reasons why gender matters" - the whole pamphlet

Looking closer at the "21 reasons why gender matters" issue, apparently the NARTH page I linked to earlier was just the summary of the main points. The full text is much more detailed, and is available online. Like most anti-gay tracts, it's brimming with footnotes, based on the usual anti-gay assumption that pointing to the sheer volume of footnotes is a good enough substitute for actually examining their quality and accuracy. Seriously. One of the first footnotes (footnote 2 to be precise) doesn't really qualify as an academic footnote at all, being just an exhortation to "refer to section 11" of the exact same document.

The pamphlet has its own dedicated site here. I'll give the footnotes a stab if I have time.

The conservative movement's mythology versus statistical evidence

One of the recurring themes of the McCain-Palin campaign has been that Obama is "out of touch", or "doesn't understand the problems of everyday Americans"

Well, check out this poll, and understand the difference between electioneering and observable reality.

By the by, I've settled on the phrase "conservative moment" to try and distinguish what I see as two conflicting versions of conservative ideology. The first is "classical conservatism", which values respect for tradition, opposes radical change and is, although not my preferred intellectual preference, a fairly useful philosophy to have around. The other, "movement conservatism", is little more than an angry mob railing against anything slightly different to them, and which is coming dangerously close to turning into a movement of brownshirts. Both currently exist in today's Republican party I think. But the number of non-movement conservatives in it appears to be dwindling rapidly.

Running the numbers: Obama/McCain polling and changes in their support

At the time of writing, The American Research Group has provided some statistics on the American Presidential election. What grabbed my attention was the detailed breakdown of some of the numbers. Now, it's possible to tie yourself into all sorts of ridiculous knots with this kind of demographic detail with questions about how Candidate A is faring against Candidate B amongst female immigrant blue-collar dog owners. Nevertheless, this is eye-catching:

National
Likely voters	Aug 30- Sep 1	Sep 13-15	Change (pts)
McCain:
Democrats	5%	19%	+14
Republicans	83%	84%	+1
Independents	43%	43%	0
Obama:
Democrats	88%	75%	-13
Republicans	11%	9%	-2
Independents	45%	49%	+4

This is the breakdown of support for each candidate among Democrats, Republicans and Independents, respectively. Note the striking swing among Democrats away from Obama and towards McCain. There has been a slight swing away by Republicans as well, but not a huge amount, and Obama's gained some Independent support (although McCain hasn't lost any). But it's that 2-digit swing towards McCain from people who identify as members of Obama's own party where Obama's hurting most, to my amateur eye.

The cause? Difficult to say, but it's tempting to view it at as a gender issue. After all, the same period shows Obama losing 5% of women and McCain gaining 7%. But that's ALL women, not just Democrats. Nor is it right to assume that the Democratic support Obama's lost is mostly female, although it very well could be.

Both surveys were taken after Obama had officially accepted the nomination instead of Hillary Clinton, so that alone doesn't explain it. The first survey was taken after Sarah Palin was announced as VP for McCain, but before she gave her convention speech. That's a possible area for analysis. But given the two week gap - a very eventful two weeks - between surveys, it's difficult to pinpoint the exact issue. More data is required. Unfortunately, I suspect that the required data is more detailed than most survey companies are willing to put up for free on their websites.